BROADBAND COMMUNITIES is the leading source of information on digital and broadband technologies for buildings and communities. Our editorial aims to accelerate the deployment of Fiber-To-The-Home and Fiber-To-The-Premises.
Issue link: https://bbcmag.epubxp.com/i/281446
MARCH/APRIL 2014 | www.broadbandcommunities.com | BROADBAND COMMUNITIES | 25 Cedar Falls could have followed the cable pattern, it made the decision to upgrade to fber rather than to DOCSIS 3.0. For CFU, going directly from HFC to gigabit fber was a better long-term fnancial strategy. "It's true that you can build a less expensive DOCSIS 3.0 network and deliver 20 or 30 Mbps to customers, which some are doing," states Betty Zeman, marketing manager of CFU. "If you're doing four-channel bonding, you can deliver 100 Mbps capacity to neighborhoods, plus you can control the number of customers per node to help individual customers get better speed performance. But as we analyzed the situation in 2009, customers' needs were growing too fast for DOCSIS to be a viable option, even with version 3 evolving. We'd constantly be playing catch-up with customers' demands and constantly changing the physical network to make those adjustments. "Now that we've fully upgraded to fber with gigabit capacity, all we have to do to increase an individual customer's speed is to change the customer's service tier while they're on the phone. Furthermore, we linked our infrastructure through three cities to the three largest network providers in the world, giving us multiple options to channel trafc to the public Internet. If you have only one path to an Internet backbone, you're just one errant backhoe away from a systemwide outage." A GIG FROM THE GET-GO? Once a community makes the decision to upgrade to a fber network, it faces the decision that Cedar Falls, Chattanooga, Wilson, N.C., and others grappled with: Do they roll out incrementally faster services over time (25 Mbps, 40 Mbps and so forth) until at some time in the future they ofer a gigabit, or do they roll out gigabit service from the outset? Chattanooga and Wilson, among others, introduced higher speeds incrementally. Tere are two important factors to consider: the cost of backhaul and the cost of getting fber to, and providing services for, the premises. Robert Houlihan, CTO and director of communications for CFU, says broadband project leaders have to address the cost of handing of data trafc to be backhauled to the public Internet. "Tat handof point can be very expensive and is a reason some providers are worried about going to a gigabit. However, if you already deliver 15 Mbps or higher as your base package, you won't see much of an increase in your total upstream backhaul by introducing a gigabit." Calix, a leading provider of broadband communications access systems and software, concurs. "We have now deployed several gigabit networks, including in Cedar Falls, and we are seeing no incremental increase in backhaul bandwidth," says David Russell, Calix solutions marketing director. "In fact, the networks function more efciently because subscribers get their application and data faster." Still, backhaul remains an unknown. Buy too much of it, and the network will have difculty breaking even because it is paying a lot for unused bandwidth. Buy too little or fail to ensure redundancy, and the network won't be able to meet customer demand. Oversubscription (selling more bandwidth than a network could support if all subscribers went online simultaneously) is part of sound revenue management. Network operators that are just starting out can lack the experience to know where the balance is between too much and too little backhaul speed. Ofering incremental speed increases helps minimize losses during this learning process. Managing the costs of delivering broadband to customers' premises can also be tricky when communities ofer a gig right out of the gate. "Some communities have chosen to use a network architecture called active Ethernet, in which one [dedicated] fber cable is run between each premises served and the central ofce," states Houlihan. "Tough signifcantly costlier than the alternative because you need more fber cables and electronics, active Ethernet ensures that the operator doesn't oversubscribe." Cedar Falls took an alternative route by using a GPON architecture. GPON allows a confguration similar to HFC, with splitter cabinets replacing HFC's optical nodes. A distribution fber line is run deep into a neighborhood and then split to reach up to 32 homes. Using GPON in a citywide buildout requires fewer fber lines to carry data to the central ofce compared with active Ethernet, yet oversubscription rates are a fraction of what they would be with HFC. WHAT'S RIGHT FOR YOU CFU's decision to launch its fber-to- the-home network with a gigabit option among its service oferings came as the result of extensive customer research, technology evaluations and industry observation. Te utility discovered, as other communities have, that using a one-size-fts-all approach is dangerous. Technology and its costs are often the main factors that sway the decision. Te experiences of the communities that pursued incremental increases indicate that the higher front-end expense of a launching a gigabit network is justifed by better fnancial and economic development outcomes in the long term. But logic and fscal reality do not always sing from the same hymnal. It is difcult to see past the front-end price tag while staring down those who hold the purse strings. To overcome the resistance to gigabit sticker shock, broadband champions must move the discussion from one of speeds, feeds and gigabit envy. Looking at the bigger picture of economic impact on a community, ofering a gigabit from the start makes even more sense. "A gigabit gives us a way to show businesses looking for a new location that we are committed to staying in front of their needs," states Zeman. "It tells companies, 'You can be certain that whatever your needs, you don't have to move your business to grow your business.' We get the opportunity to compete for businesses we otherwise might lose." Ultimately, Cedar Falls determined that, for what it wanted to achieve relative to its needs, this decision was the right one. v Craig Settles is a community broadband industry analyst, a strategy consultant and the host of the Gigabit Nation radio talk show. Reach him at craig@cjspeaks.com. BBC_Mar14.indd 25 3/14/14 2:46 PM